library(kableExtra)
suppressMessages(library(plotly))
library(dplyr)
library(DT)
library(stringr)
suppressMessages(library(ggplot2))
suppressMessages(library(dplyr))
library(tidytext)
library(tidyr)
library(xlsx)
library(reshape2)
#library(usmap)
#library(maps)

Introduction - What is the prevalence of diabetes in the U.S.?

For this project, I will:
1. Map out the number of amputations for diabetetics by state
2. Determine what state has the highest # of amputations and which gender is it predominately for.
3. Determine if across the U.S. if amputations are a predominately higher for a specific gender.

1. The Data

Initially, I wanted to do an analysis on diabetic amputations and socioeconomic factors. However, I was unable to obtain a large portion of the data due to HIPAA laws. I then turned to the basic question of - what is the prevalence of diabetes in the U.S.? In doing so, I found some alarming information regarding occurrences of diabetic amputations by gender.

2. Get the data and load it into a dataframe

diabetes <-read.csv("https://raw.githubusercontent.com/nilsabermudez/607/master/complications_hospitalizations_99.csv",   header=TRUE, sep=",")

3. Subset the Data by extracting the columns needed for this project.

CaData <-  diabetes[c(2,3,4,6,26)]
#CaData
CaData2 <-subset(CaData,age==0)
CaData3 <-subset(CaData2,sex==0)
#CaData3

4. Cleanse the data - remove N/A’s

Completerecords <- na.omit(CaData3)
#Completerecords
p <- ggplot2::ggplot(data=Completerecords,  aes(x=state, y=lea_count, color = year,group=year)) +
 
   geom_line(show.legend=TRUE) +   
 theme(legend.position = "top") +
 
  ggplot2::labs(x="States",y="Number of Amputations") +  ggtitle("Amputations across the U.S. due to Diabetes")+
  ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 90, vjust = 0.35, hjust=.5))   


plotly::ggplotly(p,colors = 'YlOrRd' )

5. Here we see that in 2014, Texas had the most amputations at an alarming number of 11,488 cases.

6. Let’s drill down and see what the gender distribution was for that year and that state:

Texas <- subset(CaData, state=="TEXAS", select = c("state","year", "sex", "age","lea_count"))
TexasMF <- subset(Texas, state=="TEXAS", select = c("state","year", "sex", "age","lea_count"))
Texas <- Texas[Texas$sex %in% c(1, 2),]
Texas <- subset(Texas, age==0, select = c("state","year", "sex", "age","lea_count"))

p<-ggplot(data=Texas, aes(y=lea_count, x=factor(sex))) +
  geom_bar(stat="identity", width=0.7, fill="steelblue",show.legend=TRUE)+
 theme(legend.position = "top") +
    ggplot2::labs(x="Amputations by Gender",y="Number of Amputations") +  ggtitle("2014 - State of Texas Amputations by Gender")+ 
   scale_x_discrete(labels=c("1" = "Males", "2" = "Females")) 

 p + guides(color = guide_legend(reverse = TRUE))

7. Let’s map out the U.S. Amputations by Gender for the 2014 year

AllData <- subset(CaData, age==0, select = c("state","year", "sex", "age","lea_count"))

USAmputations <-subset(AllData, year==2014, select = c("state","year", "sex", "age","lea_count"))
USAmputations <-  USAmputations[USAmputations$sex %in% c(1, 2),]
#USAmputations
p <- ggplot2::ggplot(data=USAmputations,  aes(x=state, y=lea_count, group=state, fill= sex, color=sex))  +geom_bar(stat = "identity", position = "dodge", width=.5)+theme(legend.position = "none") +
   scale_x_discrete(labels=c("1" = "Males", "2" = "Females")) +
  ggplot2::labs(x="States",y="Number of Amputations") +  ggtitle("Amputations across the U.S. due to Diabetes by Gender - 2014")+
  ggplot2::theme(axis.text.x = ggplot2::element_text(angle = 90, vjust = 0.35, hjust=.5)) +
  theme(panel.grid.major.x = element_blank()) 

 ggplotly(tooltip = c("text", "x", "fill", "y"))

8. Looking at the Poverty estimate for the state of Texas, it is over 4 million.

CensusPoverty <-read.csv("https://raw.githubusercontent.com/nilsabermudez/607/master/est14us.csv",skip=5,  header=TRUE, sep=",")
CensusPoverty <- subset(CensusPoverty,  select = c(Postal.Code, Poverty.Estimate..All.Ages))
datatable(CensusPoverty, options = list(pageLength = 5))

Conclusions

In this project:

  1. In 2014, Texas had the highest number of amputations.
  2. Males are getting amputations at a higher rate than females not only in Texas in 2014 - but across the U.S. in general.
  3. Socioeconomic challenges may be a factor for the high number of amputees in Texas.

This analysis could be presented to the diabetic community as a teaching example of the dangerous complications due to having diabetes.

References:

https://www.census.gov/data/datasets/2014/demo/saipe/2014-state-and-county.html
https://nccd.cdc.gov/CKD/default.aspx